Step by Step Installation of OEM 12c

Few important things to be consider before going for OEM 12c on your environment. First EM 12c is more resource demanding as compared to 10g & 11g versions. Second It’s re-engineered for better monitoring and administration capabilities and not compatible with older versions of agents, so the day you rolling 12c in your environment, you have to install 12c agent on all targets.

Package Requirements for Oracle Management Service

In our case we are setting up EM 12C on Oracle Linux 5.x 64 bit

binutils-2.17.50.0.6

gcc -4.1.1

glibc-common-2.3.4

glibc-devel-2.5-49 (32-bit version as well as the 64-bit version)

libaio-0.3.106

libstdc++ -4.1.1

libXtst-1.0.1-3.1(x86_64)

make-3.81

rng-utils-2.0

setarch-1.6

sysstat-5.0.5

xorg-x11-utils     (This package is required only for GUI-based interactive installation, and not for silent installation)

Read more of this post

Advertisements

GI (Oracle Restart) Upgrade from 11.2.0.1 to 11.2.0.3

Recently I have upgraded Single node GI (Oracle Restart) from 11.2.0.1.0 to 11.2.0.3.0 and thought to share with you. Though it’s not something very interesting to share yet not that bad at all.J

In this blog I’ll use GI or Grid Infrastructure, which is also Oracle Restart in our case. Here are details of GI (Oracle Restart):

Grid Infra Version: 11.2.0.1.0

Grid Home: /u01/app/oracle/product/11.2.0/grid

Host Name: mask11g

Storage: ASM

DB Version: 11.2.0.1.0

Purpose: Upgrade GI from 11.2.0.1.0 to 11.2.0.3.0

Path for upgrade: Out of place (New location on same server) followed by removal of old GI home.

Going forward from 11gR2 oracle recommend to go with out of place upgrade though in place upgrade is still available. I tried both and both are almost similar except the few minor things. In place upgrade enables you to upgrade an existing installation of GI into the same directory by replacing the existing installation files. The patch set application requires more downtime and is not recommended. This upgrade type requires less disk space.

For out of place upgrade we need almost 5 GB extra space during upgrade and we would be able to release space by removing the old grid software.

In order to make it easily understandable as a step by step approach I am attaching snapshots of upgrade process.

Software is available on MOS with patch number 10404530 and could be easily downloadable. This patch contains total 7 files but we need only p10404530_112030_platform_3of7.zip (933 MB) to complete GI upgrade. This patch is actually full release so it could be used to perform new installation as well as upgrade from old versions.

I would suggest you to read the instructions from all aspects before you test the upgrade on your environment. I couldn’t share more details or else this blog looks like upgrade companion doc J.

  1. Download and unzip the p10404530_112030_platform_3of7.zip into staging location on your server.
  2. Take backup of existing environment
  3. Upgrade the GI software
  4. Once upgrade is done then you could remove the old GI installation as it’s of no use or could leave it as it is in case you have plenty of space on your server.

Read more of this post

What happened when tablespace/database placed into backup mode?

User managed backup or physical backup without RMAN is still one of the popular method for backup and still part of mostly disk to disk backup methods (BCV or SRDF copy). Today I am going to discuss what happens when we place tablespace/database into backup mode.

Fractured Block

Unit of I/O for any system is block size and for oracle database it’s 2K/4K/8K/16K/32K while for any operating system it could be lower than that. One risk in making online backups is the possibility of inconsistent data within a block. Assume that OS I/O block size is 512 bytes while we used DB block size as 8K.

This means to copy utility of OS would take 16 I/Os to take backup of one DB block, while during the backup, DBWR could write 8K in one I/O.

So assume a scenario where copy utility is busy taking backup of 8th portion of particular block while DBWR came to overwrite same DB block (i.e. all 16 OS block pieces in one DB block). Because copy utility was still busy on taking backup of remaining blocks so it took new image from OS piece 9th onwards. From database perspective that backup copy of db block (collection of 16 OS blocks) has been fractured as its half or the OS blocks are old and remaining half are new. The result is called a fractured block, meaning that the data contained in this block is not consistent at a given SCN.

Oracle’s internal protection mechanism against fractured block

To overcome this danger Oracle introduced a protection feature which will be activate once you trigger ‘alter tablespace/database begin backup’ command. Following internal operations happened with immediate effect after triggering begin backup for tablespace Read more of this post

Pierre blog

Pierre Forstmann Oracle Database blog

flashdba

Oracle databases, storage and the high-performance world of flash memory

Future Veterans

Ramblings about Oracle

Ranjeet Srivastava

Smile! You’re at the best blog ever

Kevin Closson's Blog: Platforms, Databases and Storage

Platform, Database and Storage Topics

Real Life Database / SQL Experiences : An Oracle Blog from Vivek Sharma

Being an Oracle Professional, I like to share all my Real Life Performance Tuning Challenges and Experiences. The Content and Views on this site are my own and not necessarily those of Oracle. While, I write on my real life experiences, the resolutions mentioned are solely mine. Comments / Criticisms are always a welcome.

Frits Hoogland Weblog

IT Technology; Oracle, linux, TCP/IP and other stuff I find interesting

OraStory

Dominic Brooks on Oracle Performance, Tuning, Data Quality & Sensible Design ... (Now with added Sets Appeal)

ASM Support Guy

Just Another Crazy Oracle DBA

Exadata Certification

Just Another Crazy Oracle DBA

Carlos Sierra's Tools and Tips

Tools and Tips for Oracle Performance and SQL Tuning

Sangram keshari's Oracle Blog

The Fusion Middleware Administration & Database Administration Blog

Amit Saraswat

Just Another Crazy Oracle DBA

Oracle Scratchpad

Just another Oracle weblog

The Tom Kyte Blog

Just Another Crazy Oracle DBA

Hemant's Oracle DBA Blog

Just Another Crazy Oracle DBA

Uwe Hesse

about Database Technology

Richard Foote's Oracle Blog

Focusing Specifically On Oracle Indexes, Database Administration and Some Great Music